Optimization of diabetes prediction methods based on combinatorial balancing algorithm.

IF 4.6 2区 医学 Q1 ENDOCRINOLOGY & METABOLISM Nutrition & Diabetes Pub Date : 2024-08-14 DOI:10.1038/s41387-024-00324-z
HuiZhi Shao, Xiang Liu, DaShuai Zong, QingJun Song
{"title":"Optimization of diabetes prediction methods based on combinatorial balancing algorithm.","authors":"HuiZhi Shao, Xiang Liu, DaShuai Zong, QingJun Song","doi":"10.1038/s41387-024-00324-z","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Diabetes, as a significant disease affecting public health, requires early detection for effective management and intervention. However, imbalanced datasets pose a challenge to accurate diabetes prediction. This imbalance often results in models performing poorly in predicting minority classes, affecting overall diagnostic performance.</p><p><strong>Objectives: </strong>To address this issue, this study employs a combination of Synthetic Minority Over-sampling Technique (SMOTE) and Random Under-Sampling (RUS) for data balancing and uses Optuna for hyperparameter optimization of machine learning models. This approach aims to fill the gap in current research concerning data balancing and model optimization, thereby improving prediction accuracy and computational efficiency.</p><p><strong>Methods: </strong>First, the study uses SMOTE and RUS methods to process the imbalanced diabetes dataset, balancing the data distribution. Then, Optuna is utilized to optimize the hyperparameters of the LightGBM model to enhance its performance. During the experiment, the effectiveness of the proposed methods is evaluated by comparing the training results of the dataset before and after balancing.</p><p><strong>Results: </strong>The experimental results show that the enhanced LightGBM-Optuna model improves the accuracy from 97.07% to 97.11%, and the precision from 97.17% to 98.99%. The time required for a single search is only 2.5 seconds. These results demonstrate the superiority of the proposed method in handling imbalanced datasets and optimizing model performance.</p><p><strong>Conclusions: </strong>The study indicates that combining SMOTE and RUS data balancing algorithms with Optuna for hyperparameter optimization can effectively enhance machine learning models, especially in dealing with imbalanced datasets for diabetes prediction.</p>","PeriodicalId":19339,"journal":{"name":"Nutrition & Diabetes","volume":null,"pages":null},"PeriodicalIF":4.6000,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11324958/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nutrition & Diabetes","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1038/s41387-024-00324-z","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0

Abstract

Background: Diabetes, as a significant disease affecting public health, requires early detection for effective management and intervention. However, imbalanced datasets pose a challenge to accurate diabetes prediction. This imbalance often results in models performing poorly in predicting minority classes, affecting overall diagnostic performance.

Objectives: To address this issue, this study employs a combination of Synthetic Minority Over-sampling Technique (SMOTE) and Random Under-Sampling (RUS) for data balancing and uses Optuna for hyperparameter optimization of machine learning models. This approach aims to fill the gap in current research concerning data balancing and model optimization, thereby improving prediction accuracy and computational efficiency.

Methods: First, the study uses SMOTE and RUS methods to process the imbalanced diabetes dataset, balancing the data distribution. Then, Optuna is utilized to optimize the hyperparameters of the LightGBM model to enhance its performance. During the experiment, the effectiveness of the proposed methods is evaluated by comparing the training results of the dataset before and after balancing.

Results: The experimental results show that the enhanced LightGBM-Optuna model improves the accuracy from 97.07% to 97.11%, and the precision from 97.17% to 98.99%. The time required for a single search is only 2.5 seconds. These results demonstrate the superiority of the proposed method in handling imbalanced datasets and optimizing model performance.

Conclusions: The study indicates that combining SMOTE and RUS data balancing algorithms with Optuna for hyperparameter optimization can effectively enhance machine learning models, especially in dealing with imbalanced datasets for diabetes prediction.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
基于组合平衡算法的糖尿病预测方法优化。
背景:糖尿病是影响公众健康的重大疾病,需要及早发现以进行有效管理和干预。然而,不平衡的数据集给准确预测糖尿病带来了挑战。这种不平衡往往会导致模型在预测少数群体类别时表现不佳,从而影响整体诊断性能:为解决这一问题,本研究采用合成少数群体过度采样技术(SMOTE)和随机欠采样技术(RUS)相结合的方法进行数据平衡,并使用 Optuna 对机器学习模型进行超参数优化。这种方法旨在填补目前有关数据平衡和模型优化研究的空白,从而提高预测准确性和计算效率:首先,研究使用 SMOTE 和 RUS 方法处理不平衡的糖尿病数据集,平衡数据分布。然后,利用 Optuna 优化 LightGBM 模型的超参数,以提高其性能。在实验过程中,通过比较平衡前后数据集的训练结果来评估所提出方法的有效性:实验结果表明,增强型 LightGBM-Optuna 模型的准确率从 97.07% 提高到 97.11%,精度从 97.17% 提高到 98.99%。单次搜索所需的时间仅为 2.5 秒。这些结果证明了所提出的方法在处理不平衡数据集和优化模型性能方面的优越性:研究表明,将 SMOTE 和 RUS 数据平衡算法与 Optuna 超参数优化相结合,可以有效增强机器学习模型,尤其是在处理糖尿病预测的不平衡数据集时。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Nutrition & Diabetes
Nutrition & Diabetes ENDOCRINOLOGY & METABOLISM-NUTRITION & DIETETICS
CiteScore
9.20
自引率
0.00%
发文量
50
审稿时长
>12 weeks
期刊介绍: Nutrition & Diabetes is a peer-reviewed, online, open access journal bringing to the fore outstanding research in the areas of nutrition and chronic disease, including diabetes, from the molecular to the population level.
期刊最新文献
Late eating is associated with poor glucose tolerance, independent of body weight, fat mass, energy intake and diet composition in prediabetes or early onset type 2 diabetes. Association of ultra-processed food consumption with cardiovascular risk factors among patients with type-2 diabetes mellitus. Gestational diabetes exacerbates intrauterine microbial exposure induced intestinal microbiota change in offspring contributing to increased immune response. Soluble receptors for advanced glycation endproducts are predictors of insulin sensitivity and affected by weight loss. Impaired brain glucose metabolism in glucagon-like peptide-1 receptor knockout mice.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1